Just a side note I heard acpi and the rest was broken in some of the 2.6 kernels. The reason main reason is when I turned them off my machine does not power-off but does when they were turned on in 2.4 kernel. I can't have them on in 2.6 as I have an Nforce2 chipset and it does not work.
Tony. -----Original Message----- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] Behalf Of Tom Brinkman Sent: Sunday, March 28, 2004 5:01 PM To: [EMAIL PROTECTED] Subject: Re: [newbie] 10 CE Shutdown On Sunday 28 March 2004 07:28 am, Margot wrote: > Tom Brinkman wrote: > > On Thursday 25 March 2004 02:03 am, Tony S. Sykes wrote: > >>>http://www.iki.fi/tmb/Cooker/kernel-tmb-2.6.4-1.tmb.3mdk-1-1 > >>>m dk.i586.rpm > >>>http://www.iki.fi/tmb/Cooker/kernel-tmb-2.6.4-1.tmb.3mdk-1-1 > >>>m dk.nosrc.rpm > >> > >>Well, some success... I managed to download and install the > >> new kernel (without getting cut off this time), and the > >> shutdown now goes past the original hang point, but now it > >> gets to this: > >> > >>md: md0 switched to read-only mode. > >>Power down. > >> > >>...but it doesn't actually switch off! > >> > >>I assume that it is safe to just hit the power switch at this > >>point, but is there some setting I could change so it will > >>switch off automatically? > >> > >>Margot > > > > No sooner did I post the link to Thomas' kernel, than he > > released an updated version 2.6.4-1.tmb.4mdk > > Well, I've now experimented with several kernels, and the one > which works best so far is 2.6.4-1.tmb.3mdk - machine still > doesn't actually switch off, but everything else seems pretty > stable. > > Perhaps now is the time to look at the acpi and apic settings > that Tony suggested. I've seen these mentioned in other > threads, but mainly relating to laptops. Mine's a desktop > machine. I believe the general trend with 2.4.x kernels would be to disable apic and/or acpi. With 2.6.x and newer hardware, I believe it'll be to make sure they're both enabled. The only exception to that I currently know of is nForce chipset systems which don't handle them properly. You've got a VIA chipset which will. > Are there any standard settings for acpi and apic for different > types of machine, or is is just a question of trying various > combinations until I find one that works? Yep, trial'n error > > I know that the settings appear in the append line in > lilo.conf, but is this where I actually edit them, or is there > another file for the settings which transfers the details into > lilo.conf? No, edit lilo.conf, and try not to forget to run 'lilo' to write the changes before you reboot. (we all do sometimes ;) > > Margot, the behavior you describe is probly a hardware or > > bios configuration deficiency. Specially if you close all > > applications you don't want to start on the next boot, open > > a term and su to root, and type 'halt'. You might also > > experiment with 'reboot'. In any event, yes, you're probly > > OK to use the power switch, particularly if you don't see any > > file system warnings on the next boot. > > > > On an ATX motherboard, try either "4 sec delay" or > > "instant off" bios settings to see if one works better than > > the other. > > Is ATX a brand name, or is it some sort of specification? > Mine's a Shuttle Spacewalker MV42N - I don't want to start > fiddling with bios settings until I have some idea of what I'm > doing! > > Margot A specification. Your Shuttle is a micro ATX. ....and the system is up to date enough, that I believe with 2.6.x kernels, you should (even need/want to) enable apic and acpi. You do so by removing any 'noapic', 'nolapic', and 'acpi=off' (or acpi-=ht) from the 2.6.x kernel stanza in your lilo.conf. These features are built into the kernel, no need to explicitly enable them. With those option enabled, post the results of (should look somethin like mine): tom # cat /var/log/dmesg | grep -i acpi BIOS-e820: 000000001fffc000 - 000000001ffff000 (ACPI data) BIOS-e820: 000000001ffff000 - 0000000020000000 (ACPI NVS) ACPI: RSDP (v000 ASUS ) @ 0x000f62a0 ACPI: RSDT (v001 ASUS A7V600 0x42302e31 MSFT 0x31313031) @ 0x1fffc000 ACPI: FADT (v001 ASUS A7V600 0x42302e31 MSFT 0x31313031) @ 0x1fffc0b2 ACPI: BOOT (v001 ASUS A7V600 0x42302e31 MSFT 0x31313031) @ 0x1fffc030 ACPI: MADT (v001 ASUS A7V600 0x42302e31 MSFT 0x31313031) @ 0x1fffc058 ACPI: DSDT (v001 ASUS A7V600 0x00001000 MSFT 0x0100000b) @ 0x00000000 ACPI: Local APIC address 0xfee00000 ACPI: LAPIC (acpi_id[0x00] lapic_id[0x00] enabled) ACPI: LAPIC_NMI (acpi_id[0x00] high edge lint[0x1]) ACPI: IOAPIC (id[0x02] address[0xfec00000] global_irq_base[0x0]) ACPI: INT_SRC_OVR (bus 0 bus_irq 0 global_irq 2 dfl edge) ACPI: INT_SRC_OVR (bus 0 bus_irq 9 global_irq 9 low level) ACPI: PM-Timer IO Port: 0xe408 Using ACPI (MADT) for SMP configuration information ACPI: Subsystem revision 20040311 ACPI: Interpreter enabled ACPI: Using IOAPIC for interrupt routing ACPI: PCI Interrupt Link [LNKA] (IRQs 3 4 5 6 7 9 10 *11 12) ACPI: PCI Interrupt Link [LNKB] (IRQs 3 4 5 6 7 9 10 11 12) ACPI: PCI Interrupt Link [LNKC] (IRQs 3 4 5 6 7 9 *10 11 12) ACPI: PCI Interrupt Link [LNKD] (IRQs 3 4 5 6 7 9 10 11 12) ACPI: PCI Interrupt Link [LNKE] (IRQs 3 4 *5 6 7 9 10 11 12) ACPI: PCI Interrupt Link [LNKF] (IRQs 3 4 5 6 7 *9 10 11 12) ACPI: PCI Interrupt Link [LNKG] (IRQs 3 4 5 6 7 9 10 *11 12) ACPI: PCI Interrupt Link [LNKH] (IRQs 3 4 5 6 7 9 10 11 12) ACPI: PCI Root Bridge [PCI0] (00:00) ACPI: PCI Interrupt Routing Table [\_SB_.PCI0._PRT] ACPI: PCI Interrupt Routing Table [\_SB_.PCI0.PCI1._PRT] PCI: Using ACPI for IRQ routing PCI: if you experience problems, try using option 'pci=noacpi' or even 'acpi=off' apm: overridden by ACPI. ACPI: (supports S0 S1 S4 S5) I'm usin kernel 2.6.4-1.tmb.5mdk. As you might see from my results, both ACPI and APIC are involved with (even needed) for proper IRQ handling. If disabling either or both solves system problems .... then it's the system (hardware) that is the problem. I don't think yours is. I've read very good things about those little Shuttle systems, and they certainly look like good capable hardware choices to me. Also, 2.6.x kernels, newer the better, have increasingly capable and good support for apic and acpi that 2.4.x an older didn't. -- Tom Brinkman Corpus Christi, Texas Proud to be an American -+-+-+-+-+-+-+-+-+ Business Computer Projects - Disclaimer -+-+-+-+-+-+-+-+-+- This message, and any associated attachment is confidential. If you have recieved it in error, please delete it from your system, do not use or disclose the information in any way, and notify either the sender or [EMAIL PROTECTED] immediately. The contents of this message may contain personal views which are not necessarily the views of Business Computer Projects Ltd., unless specifically stated. Whilst every effort has been made to ensure that emails and their attachments are virus free, it is the responsibility of the recipient(s) to verify the integrity of such emails. Business Computer Projects Ltd BCP House 151 Charles Street Stockport Cheshire SK1 3JY Tel: +44 (0)161 355-3000 Fax: +44 (0)161 355-3001 Web: http://www.bcpsoftware.com
____________________________________________________ Want to buy your Pack or Services from MandrakeSoft? Go to http://www.mandrakestore.com Join the Club : http://www.mandrakeclub.com ____________________________________________________
